Whispers of Sound: Enhancing Information Extraction from Depression Patients' Unstructured Data through Audio and Text Emotion Recognition and Llama Fine-Tuning DOI
Lin Gan,

Xiaoyang Gao,

Yifan Huang

и другие.

Journal of medicine and health science., Год журнала: 2025, Номер 3(1), С. 28 - 33

Опубликована: Март 1, 2025

Mental health issues present significant global challenges, affecting over 20% of adults at some point in their lives. While large language models have shown promise various fields, application mental remains underexplored. This study assesses how effectively these can be applied to health, using the DAIC-WOZ text datasets and RAVDESS audio datasets. Given challenges missing non-verbal cues ambiguous terms data, data was incorporated during training address gaps. integration enhanced models' ability comprehend, extract, summarize complex information, particularly depression assessments. Additionally, technical optimizations, such as increasing model's max_length 8192, reduced GPU memory usage by 40%-50% improved context processing, leading substantial gains handling data.

Язык: Английский

Enhancing Community Detection With Hybrid Quantum-Inspired Evolutionary Algorithms DOI
Akshat Gaurav,

A. K. Katiyar,

Brij B. Gupta

и другие.

IGI Global eBooks, Год журнала: 2025, Номер unknown, С. 221 - 244

Опубликована: Фев. 28, 2025

The chapter expands upon the concepts of using hybrid QIEAs in enhancing existing community detection networks. Thus, as communities are becoming more digital untold need for sophisticated algorithms that can handle and coordinate interactions is critical. In this chapter, reader will be introduced to general quantum computing an overview technique evolutionary algorithms, which allow subsequent detailed elaboration interconnection between two fields. we first introduce idea development usage QIEAs, focusing on theoretical methods integrate classical information solving problems cannot approached by traditional EA.

Язык: Английский

Процитировано

0

Whispers of Sound: Enhancing Information Extraction from Depression Patients' Unstructured Data through Audio and Text Emotion Recognition and Llama Fine-Tuning DOI
Lin Gan,

Xiaoyang Gao,

Yifan Huang

и другие.

Journal of medicine and health science., Год журнала: 2025, Номер 3(1), С. 28 - 33

Опубликована: Март 1, 2025

Mental health issues present significant global challenges, affecting over 20% of adults at some point in their lives. While large language models have shown promise various fields, application mental remains underexplored. This study assesses how effectively these can be applied to health, using the DAIC-WOZ text datasets and RAVDESS audio datasets. Given challenges missing non-verbal cues ambiguous terms data, data was incorporated during training address gaps. integration enhanced models' ability comprehend, extract, summarize complex information, particularly depression assessments. Additionally, technical optimizations, such as increasing model's max_length 8192, reduced GPU memory usage by 40%-50% improved context processing, leading substantial gains handling data.

Язык: Английский

Процитировано

0